Global Free Energy Blog

Well this is interesting. The homemade diode current just tested a few minutes ago has made a jump to 20.6pA. Is this caused by an “energy jump?”  This jumping to various levels has been been observer numerous times in diode experiments, but has always settled down at ~ 10pA.

It should be pointed out that there was a change in the setup where the element has not been shorted, but connected to the 165 Mohm load. It is now shorted again.

It’s possible the homemade diode still has a small amount of remaining *stored* energy, and releases this above the 10pA constant by jumping to the next energy level, 20pA. And perhaps when this stored energy is released, it will drop back down to 10pA.

I don’t view the 10pA of current (@ whatever voltage) that occurs over time as a release of *stored* energy. It’s a stream of energy that seems to come from an endless reservoir beyond the component / element.

The following experimental updates occurred over the past several days, but to sum it all up, it appears that the *main* effect is caused by the contact of dissimilar materials, the same effect that produces the diode. When two dissimilar materials come into contact, electrons flow to one side, forming an electric field (same as the diode). It is believed that this electric field causes an effect that results in DC current & voltage. This would be an unknown effect. It is still unknown if the magnetic field contributes to the net DC current & voltage.

Common white paper that is soaked with water: 0.75 volts

Wax paper: 0.50 to 0.54 volts

Wax paper, flipped (in case it’s only sided wax paper): 0.50 to 0.54 volts

With transparent plastic, relatively thick: 0.36 volts

All of these recent measurements were done with an electrometer.

It’s a guarantee that soaked wet paper will produce a lot of electrochemical reactions, but if this is electrochemical then I would have expected the experiment using plastic as the insulator to produce no measurable voltage. For example, I have some extremely dead 1.5V Alkaline batteries that produce a few milli volts.

[2009/10/26 note: It's now known that the more the materials are disturbed (e.g., from measurements) the lower the voltage drops. Therefore the 0.36 volts measured where plastic was used as the insulator is most likely considerably higher if it was taken before the other measurements. It was difficult to notice if the voltage dropped much when flipping the wax paper, as the voltage was fluctuating a lot at the time. Although I did see a gradual decline in most of such experiments.]

---

What’s interesting about diodes & piezos is that regardless of how many are placed in parallel, they still produce 10pA. Well, that is, when the diode or piezos have stabilized. Initially they seem to store charge, far more than their capacitance, so after that’s been discharged, they will settle down near 10 pA.

Maybe the magnet experiments will be the same, after being shorted long enough, they will produce 10pA.

I think the soaked wet white paper was mostly electrochemical reactions. The wax paper had some electrochemical reactions. And I’d expect no measurable reactions in the plastic. So where is this 0.36 volts coming from? Maybe from metal to metal reactions??? BTW, the metal & plastic was cleaned.

---

What’s interesting is the magnet experiment with plastic as the insulator that was producing 0.36 volts yesterday is now dead. So far that’s showing the same characteristics as diodes & piezos. That is, they are easily disturbed, especially diodes. So much so that even taking too many measurements with a voltage meter can place them in the disturbed state for a long time.

---

A metal steel bolt was placed against a disc NdFeB magnet, separated by paper. Here are the measurements:

It started out producing 100mV, but continued to slowly *increase*. It went up to 150mV, and then began slowly declining. Okay, so far this is a matches diode and piezo experiments. The EE, by profession, who connected his electrometer to a data logger (months ago), saw the same thing where the voltage slowly increased, and then it slowly decreased.

After reaching the 150mV peak, it began to slowly decrease. I moved the probe wires that are connected to the magnets around to different locations. The voltage slightly jittered around a bit, nothing unexpected, but didn’t make much difference. I then took everything apart, separated the magnets. Put it back together except in *different locations* on the paper (fresh areas in case it’s due to electrochemical reactions), and the voltage was ~ 2mV; i.e., dead. It appears to be “disturbed.”

So maybe the voltage is not due to the paper. Moving it to different locations on the paper would not change nothing. Or maybe the chemicals in the magnet were drained. Fine, so I flipped the magnets to the other side. So it’s using the other side of magnet, which is an unused surface. No change! The voltage was also low. It’s as if handling the magnets and/or paper disturbes then, which is also seen in diode & piezo experiments. There are a lot of ways to disturb the diodes– rapid temp changes, appreciable current. If the diode casing is transparent, then shining low light levels on it could easily place it in the disturbed state for a long time.

I then cleaned the magnets, and tried it again. Same thing, low voltage.

There’s something mysterious going on that so far matches the diode & piezo experiments. The only two final experiments is to see if it recovers when left undisturbed, and see if the current will drop & settle to 10pA when shorted long enough.

The NdFeB magnets are 0.47″ diameter, 0.12″ thick, and have a metal coating that’s probably the typical nickel. The white paper thickness is 5 mills.

---

Here’s another experiment using large Ferrite magnets instead of NdFeB magnets. The same type of 5 mill white paper was used, and used electrometer to measure voltage,

First, 30mV. Remained relatively constant except with a *slow* decline.

Reversed the clips to see if the voltage flipped: -10mV.  Remained relatively constant except with a *slow* decline.

Reversed again: 9mV.  Remained relatively constant except with a *slow* decline.

Reversed again: -8mV.  Remained relatively constant except with a *slow* decline.

Reversed again: 2mV.  Remained relatively constant except with a *slow* decline.

As you can see, the polarity reversed, and with each reversal of the probe clip leads the voltage reversed, and also the voltage decreased by noticeable amounts each time. So far this matches the diode & piezo behavior.

My best wild guess is that the NdFeB PM’s are ~~ 5 Tesla, and the Ferrite magnets are ~~ 1 Tesla, but that could be way off.

The Ferrite magnets are rectangular, measuring 1.85 x 0.96″ and 0.39″ thick.

---

Two steel nuts were placed against each other without any magnets, but separated by paper. The produced voltage was ~ 1mV. There is most likely a hundred or so gauss in the steel nuts, residual magnetic field.

---

The two Ferrite magnets separated by paper is now up to 80mV. The parts sit undisturbed without the clip leads. So it’s not connected to anything. And then every so often I’ll connect the electrometer clip leads to take a quick measurement. As comparison, most diodes are so sensitive that if I took more than 1 measurement every two days that it will begin to slowly become disturbed. Piezos on the other hand seem far more powerful in that they can take a lot more abuse; i.e., measurements. We’ll have to see how the magnets hold up. So far they seem more sensitive than the piezo, but better than the diodes.

---

Part of this effect might be due to magnets, but at least a good portion of it in some of the experiments is not due to magnets. Also, I do not believe it is due to electrochemical reactions, or electrostatic (from friction), or any effect known to conventional physics, but hopefully tomorrow we’ll see.

Very important experiment:

I just finished some interesting testing of two Aluminum plates separated by white paper, no magnets. It’s important to note that the surface of one of the Al plates is definitely different. Perhaps it’s looks different because it has a rougher surface? Perhaps it has a thicker oxidized layer? Using the electrometer, the voltages can range from 100mV to over 400mV. I tried rubbing the paper and Al plates with different materials, friction, and it made no different in the voltage. Friction will produce high voltages, but it’s extremely high impedance. The resistance of the paper ranges from a few hundred mega ohms to 1/2 giga ohm, which in terms of insulation resistance is almost nothing.

Also I reversed each part, one at a time. First the paper, then each Al plate. This made no difference in the voltage polarity. I placed the Al plates in a magnetic field, and did not see any noticeable difference in voltage.

A 10Mohm resistor was placed across it, and the voltage dropped to 2.5mV, which comes to 250pA. After that, everything was turned off, and the Al plates were shorted to help decrease the time required for the device to reach a stabilized DC current.

So hopefully by tomorrow the current will have stabilized, … and I’ll be crossing my fingers that it will be the mysterious 10pA DC!  That will make my month!! If it’s the mysterious 10pA constant, then it means this is caused by the same effect as the diodes & piezos.

The piezos, and even more so the diodes have been tested extensively against known effects such as electrochemicals, which it is not. Regardless of how many diodes are placed in parallel, the current always stabilizes at 10pA DC.

---

Yesterdays experiment of using plastic instead of paper produced 0.36 volts. So it’s difficult to explain this as being electrochemcial. Also, yesterday after the voltage decayed down to almost nothing, moving all of the parts (magnet and clip leads) to a different location on the paper and also flipping the magnet to the other side (a fresh side) did *not* revitalize the voltage– the voltage remained low. If it was due to electrochemical, then moving the parts to new & fresh areas should produce the normal voltage again, which it did not.

If tomorrow it’s down to 10pA, then in all likelihood it caused by the internal electric field that is produced when to different materials come in contact. Of course this electric field does not directly produce DC current, but it is believed that this intense electric field causes an effect that results in DC current. As to why remains to be seen. It could be anything from Quantum tunneling to a ZPE effect to something completely unknown.

BTW, diodes have this intense internal electric field at the junction. Piezos have an intense electric field.

---

This morning, 2009/10/26, the current was 79pA.  Just 69 away from the mysterious 10pA. Will in make it, or fall below 10pA. I’m betting it will slowly decay to 10pA and settle there.

Continued from Green LED array measurements p3

At 2:52 pm PT, the green LED array, radio shack part number 276-009, was producing 695 mV DC, and climbing. The electrometer was turned off after the measurement because I had the dreadful thought of having to replace the batteries for the electrometer. The new updated electrometer has metal foil wrapped all around the batteries. It would take an act of congress to replace such batteries.

Between ~ 2:42 pm and ~ 2:52 pm the green array charged the 0.01 uF capacitor from 588 mV to 695 mV DC, which comes to ~ 1.8 pA. The resistance cannot be calculated since we do not know how much DC voltage the array was producing at that very moment. The the early time frame, 2:25 pm to 2:42 pm the array was produced ~ 1.1 pA. So the green array DC current is continuing to climb as it becomes less disturbed. It is expected to reach ~ 10 pA DC. The array is quickly recovery as it produces more DC current and voltage. According to experimental data, an undisturbed diode produces ~ 10 pA DC.

This green array would produce at least 1.10 V * 10 pA = 11 fW. IMO this array will produce far more than 1.1 volts when completely undisturbed. My best guesstimate is 0.5V * 6 = 3 volts, which comes to 30 fW.

Continued from Green LED array measurements p2

At 2:42 pm PT, 78 F, the green LED array, radio shack part number 276-009, was producing 588 mV DC, and climbing. It is noted that the LED array DC voltage will begin to slowly decrease when the electrometer is connected to it, which is nothing unexpected, for technical reasons that are beyond this blog post. In one sentence, the electrometer input bias current has higher precedence than the LED array, which causes a very slow disturbance to the diodes. I have left the electrometer on for the past several hours, and merely tilt it to take a measurement.

Between ~ 2:42 pm and ~ 2:25 pm the green array charged the 0.01 uF capacitor from 480 mV to 588 mV DC, which comes to ~ 1.1 pA. The resistance cannot be calculated since we do not know how much DC voltage the array was producing at that very moment.

Continued at Green LED array measurements p4

Continued from Green LED array measurements

At 1:10 pm PT the green LED array, radio shack part number 276-009, was producing 393 mV DC.

At about ~ 2:20 pm PT I decided to deliberately shine light on the array, not a terrible amount, but enough to easily see it. The DC voltage flipped, as expected. The DC voltage produced by light (photoelectric effect) from the LEDs and diodes with transparent cases (e.g., radio shack 1N914) has always been in the opposite voltage polarity as the DC voltage produced by the diode itself. The DC voltage was -240 mV, due to the photoelectric effect. After the light shined on the LED was removed, the DC voltage began to slowly reverse as the diodes reversed the charge on the low leakage 0.01 uF capacitor.

At about ~ 2:25 pm it’s 480 mV DC. It appears the LED array likes the new tilt switches that have ultra high resistance in the off position. Another option is that the LED array is so disturbed that it is oscillating as it dampens down. This has been an observed effect in highly disturbed diodes.

Continued at Green LED array measurements p3

A few minutes ago at 12:25 pm PT the 6 in-series green LED array, radio shack part number 276-009, was measured at producing 0.101 volts. This is a good sign, as it appears the diodes are recovering at a relatively fast rate, *or* the diode array was not as disturbed as initially thought. Again, the array was producing 1.10 volt until the flashlight was shined near the array, which caused the DC voltage to drop far below 100 mV.

Continued at Green LED array measurements p2

Today the 6 in-series green LED array, Radio Shack part number 276-009, shattered the record for maximum recorded DC voltage at producing 1.10 volts DC!

Here’s the details from past two days of measurements –>

Yesterday 2009/7/30, 5 pm PT, 78 F, 232 mV. This measurement was taken shortly after placing the green LED array inside the Hammond shield connected to the electrometer input. LED’s were inside dark room.

Today, 17 hours later as the LED array sat undisturbed inside a thick Hammond metal shield while connected to a low leakage capacitor, 2009/7/31, 10 am PT, 76 F, 1.10 volts

After seeing the 1.10 volts I stared at it in utter amazement for perhaps 10 seconds, but it seemed as eternity. The 1.10 volts was stable. After staring at it, I immediately opened the shield lids, grabbed a flashlight, turned it on, and shined it on the new tilt switches to verify they were making contact. Indeed, they were making contact, and the DC voltage was legit. I knew the flashlight would highly disturb the LEDs even though the light was not directed at the LEDs. Within ~ 5 to 10 seconds later I looked at the voltage meter, which showed the DC voltage dropping at a fast rate as the LEDs became disturbed, as expected.

The diode research is getting exciting! Yesterday, I made some mechanical tilt switches to replace the Mercury tilt switches because even the slightest disturbance decreases the Mercury switch off resistance far to low for such voltage measurements. The new tilt switches are made to have ultra high off resistance. Here’s how they were made. An ultra high resistance insulated metal wire, about 1″ long, arced 90 degrees, where one end had a small loop that had a short insulated wire dangling from the loop. The other end was soldered to an INA116PA electrometer input pin. The same was done to the other INA116PA input pin. The dangling wires would rotate as the entire setup was rotated. On the other side, separated ~ 1 cm, are two short insulated wires where the wire ends are split to form a V-shape. When the set up is rotated, the dangling wires make contact to the V-shape wires, which is connected to the LED array & low leakage capacitor. Her’s a quick drawing of the new tilt switches –>

[edit: I no longer recommend this exact type of homemade tilt switch unless you place an appreciable amount of weight on the hanging wire. After ~ a day a micro layer of oxide forms on the copper, which makes a poor contact. The weight of the short swinging wire is insufficient to make good metal-metal contact. This switch would work if weight was added to the swinging wire. When time permits I'll make a much improved tilt switch. For now, I'm using a *ball* tilt switch that was purchased last year-- part number is unavailable.]

The green and blue parts are electrically insulated metal wires. The black lines show the metal. The red arrows show the direction which the short green wire rotates when the setup is tilted. This allows for great separation distances, which equates to ultra high resistance.

Great care was taken to be certain that when the setup was flat (off) that the LED array was connected to exceptionally high resistance, probably well above 10 Tohm. There was no glue, no tape, or any such material between the LED array output pins. The LED array sits on a thick clean plastic plate. This allows the LED array to become fully undisturbed.

Anyhow, IMO the 6 in-series LED array would probably produce more than 1.10 volts when fully undisturbed, but unfortunately I have no idea how long it’s going to take the LEDs to become undisturbed after shining a relatively bright flashlight near them. The LEDs are coated with Liquid Paper on all sides except the bottom, which was facing the flashlight. The light had two affects on the LED array –>

1. It disturbed the LEDs.

2. It caused the LED array resistance to significantly drop, which appreciably discharged the low leakage capacitor.

So the LED array needs time to become undisturbed, and it needs to charge the low leakage capacitor again. Normally I would say this would take at least three weeks, but this will be the first time that extreme care was taken in being certain the diode array is connected to ultra high resistance. This might decrease the diode recovery time.

If twice as many LEDs were placed in series, for a total of 12, it appears they could produce enough DC voltage to charge a 0.01 uF capacitor that could flash an efficient LED enough to be perceived by a person.

One final note, yesterday the entire diode testing setup was moved from the garage to inside in the lab. The reason being is that the garage temperature extremes is significant compared to inside the lab. This is an attempt to minimize diode disturbance. Maybe a sign should be placed near the testing lab, “Shhhh, diodes are present!”

Over the past several weeks I’ve been a bit hush what’s been happening, but a lot of notable scientists already know, so it’s probably no secret now. I put the electrometer back together per request. Then went to radio shack and purchased two types of LEDs, red LED (part number 276-008), and a green LED (part number 276-009). Two arrays were built. If memory holds true, there are six red LEDs in-series, and the other array is 7 green LEDs in-series. I might have it backwards, but whatever. I knew from experience that both arrays would be exceptionally disturbed from all of the soldering. On the first day the red LED array produce 199 mV DC. The next day I built the green array, took out the red array, place in the green array. The green array produced 73 mV the first day.

So those are two part numbers from radio shack that include the DC voltage they produced on the first day. As much care as possible was placed in trying to keep as much light as possible from shining on the LEDs, even while at Radio Shack. Ask to have the LEDs placed in a bag, and then put it inside your pocket.

Here’s the log. All tests were done inside the garage, which can get relatively hot when the metal garage door is closed. –>

Red LED array:
2009/7/13, 4 pm, 199 mV
2009/7/14, 7:10 am, temp was 68.5 F, 145 mV

Green LED array:
2009/7/14, 8:40 am, removed red LED array, put in green LED array.
2009/7/14, 12:30 pm, 83.8 F, 71.0 mV
2009/7/14, 7pm, 87.3 F, 66.8 mV. Had a chance to discharge the low leakage capacitor to measure how fast the array recharges it, which came to 0.22 pA, which comes to 300 Gohm. The diodes were exceptionally disturbed.
2009/7/15, 8:20 am, 73 F, 102 mV.
2009/7/15, 6:50 pm, 84.4 F, 86.0 mV.
2009/7/16, 9:05 am, 71.4 F, 93.2 mV
2009/7/16, 1 pm, 86 F, 61 mV
2009/7/17, 8:50 am, 67.9 F, 68.4 mV. It appears diodes are becoming unstable by showing the signs of oscillating. This often results in the DC voltage making a jump in voltage, either up or down, as if there are preferred energy levels.
2009/7/18, 9:30 am, 73.9 F, +27.1 mV.
2009/7/19, 8:30 am, 71.3 F, +34.7 mV. [added after post]

The above pattern shows the common results of a slow voltage decay when taking measurements one or more times per day. When the measurements are cut back or halted for some time there is a pattern of the DC voltage slowly rising again, but this is also a slow occurrence. This demonstrates just how easily disturbed the diodes ability to produce a DC voltage at so-called thermal equilibrium.

There’s also a chance that the above diode disturbance was contributed by soldering the diodes, which could cause a slow avalanche effect.

Some good news, after letting the 9p White LED array rest the day without anything connected to it, the measured DC voltage was 117 mV. Since the electrometer is presently taken apart to get ready for the new setup, I had to once again use the AM-240 meter. The input wires were reversed, which showed an opposite polarity. During the first measurement, the DC voltage went up to 117 mV, reached a peek, and then began to slowly drop. Although I did not time anything, the rate at which it dropped was ~~ one milli volt per minute. By the end of the last reading where the input leads were reversed, the AM-240 showed -102 mV, which is 102 mV since the leads were reversed.

The predicted DC voltage on the AM-240 is 150 mV. That of course assumes this LED arrays net Rz resistance is well above 15 Gohm, which it appears to be. The measured input resistance of my particular AM-240 was ~ 15 Gohm. The datasheet simply shows > 100 Mohm. Months ago I contacted the manufacturer, and this datasheet value is simply to be conservative since it varies between meters. We can see in similar meters where the datasheet shows 1 Gohm, which is also conservative IMO. Companies will often be on the safe side to decrease the amount of law suites filed against them, as the electronic industry is very aggressive in the USA.

The predicted DC voltage of 150 mV is based on my measurement of the AM-240 at ~ 15 Gohm. I usually place the “~” symbol before the 15 Gohm, which means “roughly.”  The reason being was that it was a quick measurement since I saw no need to spend a lot of time to obtain high precision. The red LED was producing 123 mV. It is possible that my AM-240 input resistance is closer to 12 Gohm, but in all likelihood both of these diodes are slightly disturbed, where the 9p White LED is slightly more disturbed than the red LED.

Anyhow, it appears the 9p White LED array is now appreciably undisturbed. In a few moments I’ll post another blog discussing the possible consequences of this data.

The capacitor and resistor connected to the 9p White LED was just removed. It had a 45 Mohm load and a 1.0 uF low leakage capacitor.